Changes in social behavior and personality are often the first symptoms of a neurodegenerative disorder, and eventually occur in the course of most dementias. However, no tools currently exist that have been validated to measure social behavior deficits in patients who have limited cognitive capacity and endurance for neuropsychological testing. The specific aim of this project is to adapt existing measures of social cognition for reliable and valid use in neurodegenerative diseases. The resulting battery will 1) operationalize the social criteria for frontotemporal lobar degeneration to improve early diagnosis, 2) identify characteristic patterns of social function in other dementias, 3) provide a valid, normed measure of social function in healthy older adults, 4) link quantitative data about social cognition with structural neuroanatomy. A five-year cross-sectional investigation will be performed with a cohort of 325 patients (50 Probable Alzheimer's, 50 Possible Alzheimer's, 50 frontotemporal dementia, 50 cerebrovascular disease, 25 semantic dementia, 25 progressive nonfluent aphasia, 25 dementia with Lewy-bodies, 25 progressive supranuclear palsy, and 25 corticobasal degeneration), as well as 100 healthy older controls as a normative reference group. Subjects' social behavior and cognition will be measured using multiple assessment modalities such as questionnaire- based informant reports, observational clinical ratings, and face-to-face measures of general and social cognition, including computerized testing. The social cognition battery will be validated to determine tests' item discrimination, internal consistency, practice effects, alternate form reliability, interrater reliability, and 1-year test stability. We will derive reliable change indices and a normative dataset of older adults' performance on each measure. We will also use cluster analysis to determine whether cognitive deficits in non-social domains such as memory, language, etc. exert a systematic bias upon the performance of any social test in the battery. Multitrait- multimethod matrices and factor analysis will be done to determine the construct validity of the tests when used both with healthy older adults and with neurodegenerative disease patients. Predictive validity will be tested using categorical discriminant function analysis to identify the patterns of social deficits that predict each diagnosis, and by using voxel-based morphometry of structural MRIs to determine if specific social symptoms correspond to expected patterns of regional brain tissue loss. The resulting battery could be used by clinicians 1) to aid in the differential diagnosis of dementia; 2) to provide treatment interventions, prognostic information, and caregiver support based on a patient's social symptoms; 3) to quantify changes in social function over time in individual patients. Researchers could use such a battery 1) to objectively quantify social behavior changes as an outcome measure in treatment trials, 2) to investigate the biological correlates (i.e., genetics, proteomics, neuroanatomy, etc.) of social behavior in groups of patients with neurologic conditions, 3) to carefully quantify social deficits in a variety of disease groups and thoroughly characterize each disease's clinical phenotype. PUBLIC HEALTH RELEVANCE: The primary goal of this project is to establish a psychometrically validated battery of social cognition and behavior measures that can be used with both healthy older adults and patients with neurodegenerative diseases. Having the right tests to recognize and objectively measure social symptoms will directly contribute to improved diagnosis of frontotemporal lobar degeneration-spectrum diseases, and will provide a means for clinicians to better characterize their patients' symptoms and measure disease progression. This measurement tool will also provide a unified means by which researchers can study social functioning in both normal aging and dementia, perhaps clarifying brain-behavior relationships or utilizing social symptom improvement as a valuable outcome measure in treatment trials. [unreadable] [unreadable] [unreadable]